Analysis of cell velocytos with velocyto + scVelo

https://github.com/basilkhuder/Seurat-to-RNA-Velocity#generating-loom-files

import scvelo as scv
import anndata
import pandas as pd
import numpy as np
import matplotlib as plt
import scanpy as sc
import os
import re
import glob
import matplotlib

os.chdir("/scratch/cs/csb/projects/cropseq_nk/")
scv.set_figure_params()

scv.settings.verbosity = 3  # show errors(0), warnings(1), info(2), hints(3)
scv.settings.presenter_view = True  # set max width size for presenter view
scv.settings.set_figure_params('scvelo')  # for beautified visualization

Download the velocyto'd data

The problem is that velocyto is counted for individual samples. That's why we need to then also add the merged UMAPs and clusters to the merged .loom objects.

Also, from those .loom objects some of the cells did not survive the QC. That's why we need to remove them

Timepoints_plate1 = scv.read("data/velocyto/new_2023/Timepoints_plate1.loom", cache=False)
Timepoints_plate2 = scv.read("data/velocyto/new_2023/Timepoints_plate2.loom", cache=False)

Variable names are not unique. To make them unique, call `.var_names_make_unique`.
Variable names are not unique. To make them unique, call `.var_names_make_unique`.

Timepoints_plate1_expanded_sample_obs    = pd.read_csv("data/velocyto/new_2023/Timepoints_plate1Timepoints_plate1_cells.csv")
Timepoints_plate1_expanded_umap_cord     = pd.read_csv("data/velocyto/new_2023/Timepoints_plate1Timepoints_plate1_embeddings.csv")
Timepoints_plate1_expanded_cell_clusters = pd.read_csv("data/velocyto/new_2023/Timepoints_plate1Timepoints_plate1_cluster.csv")

Timepoints_plate2_expanded_sample_obs    = pd.read_csv("data/velocyto/new_2023/Timepoints_plate2Timepoints_plate2_cells.csv")
Timepoints_plate2_expanded_umap_cord     = pd.read_csv("data/velocyto/new_2023/Timepoints_plate2Timepoints_plate2_embeddings.csv")
Timepoints_plate2_expanded_cell_clusters = pd.read_csv("data/velocyto/new_2023/Timepoints_plate2Timepoints_plate2_cluster.csv")

totaltotal_sample_obs    = pd.read_csv("data/velocyto/new_2023/totaltotal_cells.csv")
totaltotal_umap_cord     = pd.read_csv("data/velocyto/new_2023/totaltotal_embeddings.csv")
totaltotal_cell_clusters = pd.read_csv("data/velocyto/new_2023/totaltotal_cluster.csv")

Subset

In this part we remove the QCd cells.

Timepoints_plate1  = Timepoints_plate1[np.isin(Timepoints_plate1.obs.index, Timepoints_plate1_expanded_sample_obs["x"])]
Timepoints_plate2  = Timepoints_plate2[np.isin(Timepoints_plate2.obs.index, Timepoints_plate2_expanded_sample_obs["x"])]

Merge

Timepoints_plate1.var_names_make_unique()
Timepoints_plate2.var_names_make_unique()

Variable names are not unique. To make them unique, call `.var_names_make_unique`.
Variable names are not unique. To make them unique, call `.var_names_make_unique`.
Variable names are not unique. To make them unique, call `.var_names_make_unique`.
Variable names are not unique. To make them unique, call `.var_names_make_unique`.

Timepoints_plate_tot = Timepoints_plate1.concatenate(Timepoints_plate2)

Add the UMAPs and clusters from the total object

tot_expanded_index     = pd.DataFrame(Timepoints_plate_tot.obs.index)
totaltotal_umap_cord = totaltotal_umap_cord.rename(columns = {'Unnamed: 0':'CellID'})
tot_expanded_index['CellID'] = tot_expanded_index['CellID'].str.slice(stop=-2)

totaltotal_umap_cord_ordered = tot_expanded_index.merge(totaltotal_umap_cord, on = "CellID")
totaltotal_umap_cord_ordered = totaltotal_umap_cord_ordered.iloc[:,1:]
Timepoints_plate_tot.obsm['X_umap']  = totaltotal_umap_cord_ordered.values

tot_expanded_index     = pd.DataFrame(Timepoints_plate_tot.obs.index)
totaltotal_cell_clusters = totaltotal_cell_clusters.rename(columns = {'Unnamed: 0':'CellID'})
tot_expanded_index['CellID'] = tot_expanded_index['CellID'].str.slice(stop=-2)
totaltotal_cell_clusters_ordered = tot_expanded_index.merge(totaltotal_cell_clusters, on = "CellID")
totaltotal_cell_clusters_ordered = totaltotal_cell_clusters_ordered.iloc[:,1:]
Timepoints_plate_tot.obs['clusters_ord'] = totaltotal_cell_clusters_ordered.values

Timepoints_plate_tot.obs

	Clusters	_X	_Y	batch	clusters_ord
CellID
Timepoints_plate1:AAAGGATCAGCGGTTCx-0	0	11.025361	-13.172872	0	Cytokine
Timepoints_plate1:AAACGCTTCCCTCAACx-0	4	47.889122	-6.541481	0	Activated
Timepoints_plate1:AAACGCTAGTTCATGCx-0	4	21.255173	0.206297	0	Early activated
Timepoints_plate1:AAAGGTACAGCAGTTTx-0	0	29.986277	-20.724464	0	Early activated
Timepoints_plate1:AAACCCAGTCACTGATx-0	5	-37.813477	-10.019014	0	Resting
...	...	...	...	...	...
Timepoints_plate2:TTTGGTTCACGGCCATx-1	7	-11.732607	-41.042072	1	Resting
Timepoints_plate2:TTTGGTTCAGACAAGCx-1	14	-18.880938	-18.124798	1	Activated
Timepoints_plate2:TTTGGTTGTCCGGACTx-1	12	-15.114505	17.845682	1	Resting
Timepoints_plate2:TTTGTTGCAGCAGTTTx-1	3	-43.133324	-12.582972	1	Resting
Timepoints_plate2:TTTGTTGCACCTTCGTx-1	4	19.752218	-35.777641	1	Type I IFN

3375 rows × 5 columns

scVelo

This is the stochastic model

Timepoints_plate_tot=anndata.read_h5ad("results/velocyto/new_2023/Timepoints_plate_tot_scv.h5ad")

scv.pp.filter_and_normalize(Timepoints_plate_tot)
scv.pp.moments(Timepoints_plate_tot)
scv.tl.velocity(Timepoints_plate_tot, mode = "stochastic")
scv.tl.velocity_graph(Timepoints_plate_tot)
scv.pl.velocity_embedding(Timepoints_plate_tot, basis = 'umap')

WARNING: Did not normalize X as it looks processed already. To enforce normalization, set `enforce=True`.
WARNING: Did not normalize spliced as it looks processed already. To enforce normalization, set `enforce=True`.
WARNING: Did not normalize unspliced as it looks processed already. To enforce normalization, set `enforce=True`.
WARNING: Did not modify X as it looks preprocessed already.
computing moments based on connectivities
    finished (0:00:06) --> added 
    'Ms' and 'Mu', moments of un/spliced abundances (adata.layers)
computing velocities
    finished (0:00:26) --> added 
    'velocity', velocity vectors for each individual cell (adata.layers)
computing velocity graph (using 1/24 cores)
    finished (0:00:31) --> added 
    'velocity_graph', sparse matrix with cosine correlations (adata.uns)
computing velocity embedding
    finished (0:00:01) --> added
    'velocity_umap', embedded velocity vectors (adata.obsm)

scv.pl.proportions(Timepoints_plate_tot, groupby = "clusters_ord", save = "results/velocyto/new_2023/bar_spliced.png")

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
/tmp/ipykernel_15795/4101793438.py in <module>
----> 1 scv.pl.proportions(Timepoints_plate_tot, groupby = "clusters_ord", save = "results/velocyto/new_2023/bar_spliced.png")

NameError: name 'Timepoints_plate_tot' is not defined

scv.tl.score_genes_cell_cycle(Timepoints_plate_tot)
scv.pl.scatter(Timepoints_plate_tot, color_gradients=['S_score', 'G2M_score'], smooth=True, perc=[5, 95], save = "results/velocyto/new_2023/umap_cell_cycle.png")

calculating cell cycle phase
-->     'S_score' and 'G2M_score', scores of cell cycle phases (adata.obs)
saving figure to file results/velocyto/new_2023/umap_cell_cycle.png

scv.pl.velocity_embedding_stream(Timepoints_plate_tot, 
                                 basis='umap', 
                                 color = "clusters_ord", 
                                 # color_map="Pastel1", 
#                                 palette = ["#2A363B”, “#FF847C”, “#99B898”, “#E84A5F”, “#96281B"],
                                 palette = ["salmon",'#CD8C95',"mediumaquamarine", "darkslategray",'maroon'],
                                 arrow_style = "-|>",
                                 size = 50,
                                 alpha = 1,
                                 linewidth = 1,
                                 dpi = 160,
                                 save = "results/velocyto/new_2023/umap_stream.png")

saving figure to file results/velocyto/new_2023/umap_stream.png

scv.tl.velocity_confidence(Timepoints_plate_tot)
keys = 'velocity_length', 'velocity_confidence'
scv.pl.scatter(Timepoints_plate_tot, 
               c=keys, 
               cmap='coolwarm',
               perc=[5, 95],
               save = "results/velocyto/new_2023/umap_confidence_length.png")

--> added 'velocity_length' (adata.obs)
--> added 'velocity_confidence' (adata.obs)
saving figure to file results/velocyto/new_2023/umap_confidence_length.png

Get the genes that are associated with the trajectories

scv.tl.rank_velocity_genes(Timepoints_plate_tot, groupby='clusters_ord', min_corr=.3)
top_scvelo_genes = scv.DataFrame(Timepoints_plate_tot.uns['rank_velocity_genes']['names'])
top_scvelo_genes.to_csv(path_or_buf = "results/velocyto/new_2023/top_scvelo_genes.csv", index = False)
top_scvelo_genes

ranking velocity genes
    finished (0:00:00) --> added 
    'rank_velocity_genes', sorted scores by group ids (adata.uns) 
    'spearmans_score', spearmans correlation scores (adata.var)

	Activated	Cytokine	Early activated	Resting	Type I IFN
0	LINC01934	ANKRD28	PHACTR2	GRAP2	CMPK2
1	AC008105.3	PTPN22	BICDL1	EML4	USP18
2	PTPRJ	NAMPT	TNIP3	ARHGAP18	TRIM22
3	DLEU2	CREM	TSPAN5	PDE7A	TRIM5
4	ZNF827	IL2RA	PPM1G	TC2N	RASGRP3
...	...	...	...	...	...
95	CENPI	NAB2	SV2C	RBL2	IFI6
96	GBP1	ICAM1	RBPJ	SYNE1	TNFSF13B
97	RAD21	TTF2	NETO2	CHEK1	SORL1
98	PRR5L	ZNF331	LARP1B	CTNNAL1	NFATC3
99	CSF1	GNG4	NME1	HP1BP3	DUSP2

100 rows × 5 columns

scv.pl.velocity(Timepoints_plate_tot, 
                ["PRF1", "GZMB", "FCGR3A", "NCAM1", 
                 "TNF", "IFNG", "CCL3", "CCL4"], 
                ncols=4, 
                color = "clusters_ord", 
                save="results/velocyto/new_2023/scatter_velocity_genes1.png")

saving figure to file results/velocyto/new_2023/scatter_velocity_genes1.png

scv.pl.velocity(Timepoints_plate_tot, 
                 ["TNF", "TNFSRF4", "TNFSRF8", "TNFRSF18", 
                 "MX1", "OAS1", "IFIT1", "IFIT3"], 
                ncols=4, 
                color = "clusters_ord", 
                save="results/velocyto/new_2023/scatter_velocity_genes2.png")

saving figure to file results/velocyto/new_2023/scatter_velocity_genes2.png

Calcualte pseudotime and PAGA-trajectories

scv.tl.velocity_pseudotime(Timepoints_plate_tot)
scv.pl.scatter(Timepoints_plate_tot, color='velocity_pseudotime', cmap='gnuplot', save = "results/velocyto/new_2023/umap_pseudotime.png")

saving figure to file results/velocyto/new_2023/umap_pseudotime.png

Write down the model

Timepoints_plate_tot.write("results/velocyto/new_2023/Timepoints_plate_tot_scv.h5ad")

top_genes1 = ["KLRC1", "NCAM1", "GZMK", "GZMA", "KLRB1", 
             "KLRC2", "GZMH", "LAG3", "HLA-DRA", "TNFRSF18", 
             "TNFRSF9", "TNFRSF4", "CRTAM", "ENTPD1", "HAVCR2",
             "TIGIT", "TNFSF10", "BCL2L11", "ISG15", "OAS1"]
top_genes2 = ["OAS2", "OAS3", "MX1", "MX2", "IRF7", 
             "IRF9", "EIF2AK2", "CCL3", "CCL4", "TNF", 
              "IFNG", "CD69"]
scv.pl.scatter(Timepoints_plate_tot, basis=top_genes1, ncols=5, frameon=False,
               color = "clusters_ord",
               palette = ['red','#CD8C95','#6E8B3D', "#D3D3D3",'darkred'],
               save = "results/velocyto/new_2023/scatter_likelohood_olli.png")

saving figure to file results/velocyto/new_2023/scatter_likelohood_olli.png

Dynamical model

Here, we use the generalized dynamical model to solve the full transcriptional dynamics. That yields several additional insights such as latent time and identification of putative driver genes.

Timepoints_plate_tot=anndata.read_h5ad("results/velocyto/new_2023/Timepoints_plate_tot_scv.h5ad")

Timepoints_plate_tot_dyn=Timepoints_plate_tot

scv.pp.filter_and_normalize(Timepoints_plate_tot_dyn, enforce=True)
scv.tl.recover_dynamics(Timepoints_plate_tot_dyn)

Normalized count data: X, spliced, unspliced.
WARNING: Did not modify X as it looks preprocessed already.
recovering dynamics (using 1/24 cores)
WARNING: Unable to create progress bar. Consider installing `tqdm` as `pip install tqdm` and `ipywidgets` as `pip install ipywidgets`,
or disable the progress bar using `show_progress_bar=False`.
    finished (0:10:09) --> added 
    'fit_pars', fitted parameters for splicing dynamics (adata.var)

scv.tl.velocity(Timepoints_plate_tot_dyn, mode='dynamical')
scv.tl.velocity_graph(Timepoints_plate_tot_dyn)

computing velocities
    finished (0:00:28) --> added 
    'velocity', velocity vectors for each individual cell (adata.layers)
computing velocity graph (using 1/24 cores)
    finished (0:00:08) --> added 
    'velocity_graph', sparse matrix with cosine correlations (adata.uns)

scv.pl.velocity_embedding_stream(Timepoints_plate_tot_dyn, 
                                 basis='umap',
                                 color = "clusters_ord",
                                 palette = ['red','#CD8C95','#6E8B3D', "#D3D3D3",'darkred'],
                                save = "results/velocyto/new_2023/umap_stream_dyn.png")

computing velocity embedding
    finished (0:00:02) --> added
    'velocity_umap', embedded velocity vectors (adata.obsm)
saving figure to file results/velocyto/new_2023/umap_stream_dyn.png

scv.tl.velocity_confidence(Timepoints_plate_tot_dyn)
keys = 'velocity_length', 'velocity_confidence'
scv.pl.scatter(Timepoints_plate_tot_dyn, 
               c=keys, 
               cmap='coolwarm',
               perc=[5, 95],
               save = "results/velocyto/new_2023/umap_confidence_length_dyn.png")

--> added 'velocity_length' (adata.obs)
--> added 'velocity_confidence' (adata.obs)
saving figure to file results/velocyto/new_2023/umap_confidence_length_dyn.png

df = Timepoints_plate_tot_dyn.var
df = df[(df['fit_likelihood'] > .1) & df['velocity_genes'] == True]

kwargs = dict(xscale='log', fontsize=16)
with scv.GridSpec(ncols=3) as pl:
    pl.hist(df['fit_alpha'], xlabel='transcription rate', **kwargs)
    pl.hist(df['fit_beta'] * df['fit_scaling'], xlabel='splicing rate', xticks=[.1, .4, 1], **kwargs)
    pl.hist(df['fit_gamma'], xlabel='degradation rate', xticks=[.1, .4, 1], **kwargs)

df = scv.get_df(Timepoints_plate_tot_dyn, 'fit*', dropna=True)
df.to_csv(path_or_buf="results/velocyto/new_2023/kin_rate_parameters_dyn.csv", index = False)

scv.tl.latent_time(Timepoints_plate_tot_dyn)
scv.pl.scatter(Timepoints_plate_tot_dyn, 
               color='latent_time', 
               color_map='gnuplot', 
               size=80, 
               save = "results/velocyto/new_2023/umap_latent_time_dyn.png")

computing latent time using root_cells as prior
    finished (0:00:19) --> added 
    'latent_time', shared time (adata.obs)
saving figure to file results/velocyto/new_2023/umap_latent_time_dyn.png

top_genes = Timepoints_plate_tot_dyn.var['fit_likelihood'].sort_values(ascending=False).index[:300]
scv.pl.heatmap(Timepoints_plate_tot_dyn, 
               var_names=top_genes, sortby='latent_time', col_color='clusters_ord', 
               n_convolve=100,
              save = "results/velocyto/new_2023/heatmap_dyn_likelohood.png")

saving figure to file results/velocyto/new_2023/heatmap_dyn_likelohood.png

top_genes = Timepoints_plate_tot_dyn.var['fit_likelihood'].sort_values(ascending=False).index
scv.pl.scatter(Timepoints_plate_tot_dyn, basis=top_genes[:15], ncols=5, frameon=False,
               color = "clusters_ord",
               palette = ['red','#CD8C95','#6E8B3D', "#D3D3D3",'darkred'],
               save = "results/velocyto/new_2023/scatter_dyn_likelohood.png")

saving figure to file results/velocyto/new_2023/scatter_dyn_likelohood.png

top_genes1 = ["KLRC1", "NCAM1", "GZMK", "GZMA", "KLRB1", 
             "KLRC2", "GZMH", "LAG3", "HLA-DRA", "TNFRSF18", 
             "TNFRSF9", "TNFRSF4", "CRTAM", "ENTPD1", "HAVCR2",
             "TIGIT", "TNFSF10", "BCL2L11", "ISG15", "OAS1"]
top_genes2 = ["OAS2", "OAS3", "MX1", "MX2", "IRF7", 
             "IRF9", "EIF2AK2", "CCL3", "CCL4", "TNF", 
              "IFNG", "CD69"]
scv.pl.scatter(Timepoints_plate_tot_dyn, basis=top_genes1, ncols=5, frameon=False,
               color = "clusters_ord",
               palette = ['red','#CD8C95','#6E8B3D', "#D3D3D3",'darkred'],
               save = "results/velocyto/new_2023/scatter_dyn_likelohood_olli.png")

saving figure to file results/velocyto/new_2023/scatter_dyn_likelohood_olli.png

scv.pl.scatter(Timepoints_plate_tot_dyn, basis=top_genes2, ncols=5, frameon=False,
               color = "clusters_ord",
               palette = ['red','#CD8C95','#6E8B3D', "#D3D3D3",'darkred'],
               save = "results/velocyto/new_2023/scatter_dyn_likelohood_olli2.png")

saving figure to file results/velocyto/new_2023/scatter_dyn_likelohood_olli2.png

scv.tl.rank_dynamical_genes(Timepoints_plate_tot_dyn, groupby='clusters_ord')
df = scv.get_df(Timepoints_plate_tot_dyn, 'rank_dynamical_genes/names')
df.head(5)
df.to_csv("results/velocyto/new_2023/top_scvelo_genes_dyn.csv", index = False)

ranking genes by cluster-specific likelihoods
    finished (0:00:04) --> added 
    'rank_dynamical_genes', sorted scores by group ids (adata.uns)

df.head(20)

	Activated	Cytokine	Early activated	Resting	Type I IFN
0	NUSAP1	GNLY	GNLY	NUSAP1	TMSB4X
1	ECT2	NUSAP1	NDC80	GNLY	NUSAP1
2	TOP2A	SMC4	TOP2A	TOP2A	EPSTI1
3	SMC4	MKI67	TMPO	NDC80	NCAPH
4	CLEC2D	CENPF	SMC4	SMC4	SMC4
5	SMC2	AURKB	NUSAP1	DIAPH3	MELK
6	RBPJ	CDK1	TMSB4X	TMPO	IFI44L
7	CRTAM	TMPO	TPX2	MKI67	EZH2
8	CDK1	LGALS1	SPC25	CIP2A	SMC2
9	MKI67	RRM2	CLEC2D	CENPE	MX1
10	TPX2	TMSB4X	EZH2	EZH2	ESCO2
11	PRC1	PRC1	HMGN2	MELK	NCAPG2
12	KIF4A	CKAP2L	ATAD2	KIF11	OSTF1
13	CKAP2L	ESCO2	CRTAM	TMSB4X	IFI6
14	MELK	OSTF1	MKI67	TACC3	KIF4A
15	TMSB4X	SPC25	DIAPH3	LNPEP	RPL5
16	NCAPG2	PTMA	CDK1	BUB1	RRM2
17	KIF11	NDC80	RBPJ	ECT2	NOP58
18	GNLY	HMGN2	ANP32E	TPX2	KIF23
19	ASPM	CENPE	AURKB	CENPF	DIAPH3

for cluster in ["Resting (0)", "Adaptive-like (1)", "Activated (2)", "IFN-stimulated (3)","Cytokine-producing (4)"]:
    scv.pl.scatter(expanded_tot_dyn, df[cluster][:5], ylabel=cluster, frameon=False,
                  save = "results/velocyto/expanded/scatter_dyn_likelihood2.png")

saving figure to file results/velocyto/expanded/scatter_dyn_likelihood2.png

saving figure to file results/velocyto/expanded/scatter_dyn_likelihood2.png

saving figure to file results/velocyto/expanded/scatter_dyn_likelihood2.png

saving figure to file results/velocyto/expanded/scatter_dyn_likelihood2.png

saving figure to file results/velocyto/expanded/scatter_dyn_likelihood2.png

Timepoints_plate_tot_dyn.write("results/velocyto/new_2023/Timepoints_plate_tot_dyn_scv.h5ad")

Cell lines as separates

K562

k562_obs              = pd.read_csv("data/velocyto/new_2023/K562K562_cells.csv")
Timepoints_plate_k562 = Timepoints_plate_tot[Timepoints_plate_tot.obs.index.str.slice(stop=-2).isin(k562_obs["x"])]

scv.tl.velocity(Timepoints_plate_k562, mode = "stochastic")
scv.pp.neighbors(Timepoints_plate_k562)
scv.tl.velocity_graph(Timepoints_plate_k562)
scv.pl.velocity_embedding(Timepoints_plate_k562, basis = 'umap')
scv.tl.velocity_confidence(Timepoints_plate_k562)
scv.tl.rank_velocity_genes(Timepoints_plate_k562, groupby='clusters_ord', min_corr=.3)
scv.tl.velocity_pseudotime(Timepoints_plate_k562)

computing velocities
    finished (0:00:01) --> added 
    'velocity', velocity vectors for each individual cell (adata.layers)
computing neighbors
    finished (0:00:00) --> added 
    'distances' and 'connectivities', weighted adjacency matrices (adata.obsp)
computing velocity graph (using 1/24 cores)
    finished (0:00:01) --> added 
    'velocity_graph', sparse matrix with cosine correlations (adata.uns)
computing velocity embedding
    finished (0:00:00) --> added
    'velocity_umap', embedded velocity vectors (adata.obsm)

--> added 'velocity_length' (adata.obs)
--> added 'velocity_confidence' (adata.obs)
ranking velocity genes
    finished (0:00:00) --> added 
    'rank_velocity_genes', sorted scores by group ids (adata.uns) 
    'spearmans_score', spearmans correlation scores (adata.var)

scv.pl.velocity_embedding_stream(Timepoints_plate_k562, 
                                 basis='umap', 
                                 color = "clusters_ord", 
                                 # color_map="Pastel1", 
#                                 palette = ["#2A363B”, “#FF847C”, “#99B898”, “#E84A5F”, “#96281B"],
                                 palette = ["salmon",'#CD8C95',"mediumaquamarine", "darkslategray",'maroon'],
                                 arrow_style = "-|>",
                                 size = 50,
                                 alpha = 1,
                                 linewidth = 1,
                                 dpi = 160,
                                 save = "results/velocyto/new_2023/umap_stream_k562.png")

saving figure to file results/velocyto/new_2023/umap_stream.png

scv.pl.scatter(Timepoints_plate_k562, color='velocity_pseudotime', cmap='gnuplot', save = "results/velocyto/new_2023/umap_pseudotime_k562.png")

saving figure to file results/velocyto/new_2023/umap_pseudotime_k562.png

Timepoints_plate_k562.write("results/velocyto/new_2023/Timepoints_plate_k562.h5ad")

GDM1

gdm1_obs              = pd.read_csv("data/velocyto/new_2023/GDM1GDM1_cells.csv")
Timepoints_plate_gdm1 = Timepoints_plate_tot[Timepoints_plate_tot.obs.index.str.slice(stop=-2).isin(gdm1_obs["x"])]

scv.tl.velocity(Timepoints_plate_gdm1, mode = "stochastic")
scv.pp.neighbors(Timepoints_plate_gdm1)
scv.tl.velocity_graph(Timepoints_plate_gdm1)
scv.pl.velocity_embedding(Timepoints_plate_gdm1, basis = 'umap')
scv.tl.velocity_confidence(Timepoints_plate_gdm1)
scv.tl.rank_velocity_genes(Timepoints_plate_gdm1, groupby='clusters_ord', min_corr=.3)
scv.tl.velocity_pseudotime(Timepoints_plate_gdm1)

computing velocities
    finished (0:00:05) --> added 
    'velocity', velocity vectors for each individual cell (adata.layers)
computing neighbors
    finished (0:00:00) --> added 
    'distances' and 'connectivities', weighted adjacency matrices (adata.obsp)
computing velocity graph (using 1/24 cores)
    finished (0:00:01) --> added 
    'velocity_graph', sparse matrix with cosine correlations (adata.uns)
computing velocity embedding
    finished (0:00:00) --> added
    'velocity_umap', embedded velocity vectors (adata.obsm)

--> added 'velocity_length' (adata.obs)
--> added 'velocity_confidence' (adata.obs)
ranking velocity genes
    finished (0:00:00) --> added 
    'rank_velocity_genes', sorted scores by group ids (adata.uns) 
    'spearmans_score', spearmans correlation scores (adata.var)

scv.pl.velocity_embedding_stream(Timepoints_plate_gdm1, 
                                 basis='umap', 
                                 color = "clusters_ord", 
                                 palette = ["salmon",'#CD8C95',"mediumaquamarine", "darkslategray",'maroon'],
                                 arrow_style = "-|>",
                                 size = 50,
                                 alpha = 1,
                                 linewidth = 1,
                                 dpi = 160,
                                 save = "results/velocyto/new_2023/umap_stream_gdm1.png")

saving figure to file results/velocyto/new_2023/umap_stream_gdm1.png

scv.pl.scatter(Timepoints_plate_gdm1, color='velocity_pseudotime', cmap='gnuplot', save = "results/velocyto/new_2023/umap_pseudotime_gdm1.png")

saving figure to file results/velocyto/new_2023/umap_pseudotime_gdm1.png

Timepoints_plate_gdm1.write("results/velocyto/new_2023/Timepoints_plate_gdm1.h5ad")

RI1

ri1_obs              = pd.read_csv("data/velocyto/new_2023/RI1RI1_cells.csv")
Timepoints_plate_ri1 = Timepoints_plate_tot[Timepoints_plate_tot.obs.index.str.slice(stop=-2).isin(ri1_obs["x"])]

scv.pl.velocity_embedding_stream(Timepoints_plate_ri1, 
                                 basis='umap', 
                                 color = "clusters_ord", 
                                 palette = ["salmon",'#CD8C95',"mediumaquamarine", "darkslategray",'maroon'],
                                 arrow_style = "-|>",
                                 size = 50,
                                 alpha = 1,
                                 linewidth = 1,
                                 dpi = 160,
                                 save = "results/velocyto/new_2023/umap_stream_ri1.png")

Trying to set attribute `.uns` of view, copying.

saving figure to file results/velocyto/new_2023/umap_stream_ri1.png

scv.pl.scatter(Timepoints_plate_ri1, color='velocity_pseudotime', cmap='gnuplot', save = "results/velocyto/new_2023/umap_pseudotime_ri1.png")

saving figure to file results/velocyto/new_2023/umap_pseudotime_ri1.png

Timepoints_plate_ri1.write("results/velocyto/new_2023/Timepoints_plate_ri1.h5ad")

Jurkat

jurkat_obs              = pd.read_csv("data/velocyto/new_2023/JURKATJURKAT_cells.csv")
Timepoints_plate_jurkat = Timepoints_plate_tot[Timepoints_plate_tot.obs.index.str.slice(stop=-2).isin(jurkat_obs["x"])]

scv.pl.velocity_embedding_stream(Timepoints_plate_jurkat, 
                                 basis='umap', 
                                 color = "clusters_ord", 
                                 palette = ["salmon",'#CD8C95',"mediumaquamarine", "darkslategray",'maroon'],
                                 arrow_style = "-|>",
                                 size = 50,
                                 alpha = 1,
                                 linewidth = 1,
                                 dpi = 160,
                                 save = "results/velocyto/new_2023/umap_stream_jurkat.png")

Trying to set attribute `.uns` of view, copying.

saving figure to file results/velocyto/new_2023/umap_stream_jurkat.png

scv.pl.scatter(Timepoints_plate_jurkat, color='velocity_pseudotime', cmap='gnuplot', save = "results/velocyto/new_2023/umap_pseudotime_jurkat.png")

saving figure to file results/velocyto/new_2023/umap_pseudotime_jurkat.png

Timepoints_plate_jurkat.write("results/velocyto/new_2023/Timepoints_plate_jurkat.h5ad")

NK-1

nk1_obs              = pd.read_csv("data/velocyto/new_2023/NK-1NK-1_cells.csv")
Timepoints_plate_nk1 = Timepoints_plate_tot[Timepoints_plate_tot.obs.index.str.slice(stop=-2).isin(nk1_obs["x"])]

scv.pl.velocity_embedding_stream(Timepoints_plate_nk1, 
                                 basis='umap', 
                                 color = "clusters_ord", 
                                 palette = ["salmon",'#CD8C95',"mediumaquamarine", "darkslategray",'maroon'],
                                 arrow_style = "-|>",
                                 size = 50,
                                 alpha = 1,
                                 linewidth = 1,
                                 dpi = 160,
                                 save = "results/velocyto/new_2023/umap_stream_nk1.png")

Trying to set attribute `.uns` of view, copying.

saving figure to file results/velocyto/new_2023/umap_stream_nk1.png

scv.pl.scatter(Timepoints_plate_nk1, color='velocity_pseudotime', cmap='gnuplot', save = "results/velocyto/new_2023/umap_pseudotime_nk1.png")

saving figure to file results/velocyto/new_2023/umap_pseudotime_nk1.png

Timepoints_plate_nk1.write("results/velocyto/new_2023/Timepoints_plate_nk1.h5ad")

NK-2

nk2_obs              = pd.read_csv("data/velocyto/new_2023/NK-2NK-2_cells.csv")
Timepoints_plate_nk2 = Timepoints_plate_tot[Timepoints_plate_tot.obs.index.str.slice(stop=-2).isin(nk2_obs["x"])]

scv.pl.velocity_embedding_stream(Timepoints_plate_nk2, 
                                 basis='umap', 
                                 color = "clusters_ord", 
                                 palette = ["salmon",'#CD8C95',"mediumaquamarine", "darkslategray",'maroon'],
                                 arrow_style = "-|>",
                                 size = 50,
                                 alpha = 1,
                                 linewidth = 1,
                                 dpi = 160,
                                 save = "results/velocyto/new_2023/umap_stream_nk2.png")

Trying to set attribute `.uns` of view, copying.

saving figure to file results/velocyto/new_2023/umap_stream_nk2.png

scv.pl.scatter(Timepoints_plate_nk2, color='velocity_pseudotime', cmap='gnuplot', save = "results/velocyto/new_2023/umap_pseudotime_nk2.png")

saving figure to file results/velocyto/new_2023/umap_pseudotime_nk2.png

Timepoints_plate_nk2.write("results/velocyto/new_2023/Timepoints_plate_nk2.h5ad")